Nest boxes are often installed in agricultural landscapes to attract barn owls for the pest control services they provide. For this practice to be effective, farmers need actionable guidelines on nest box design and placement to optimize nest box occupancy. Ongoing research has revealed nest box selection patterns in winegrape vineyards of Napa Valley, California. However, it remains unclear whether these patterns are generalizable to other years and regions with different crops. Based on 6 years of occupancy data from Napa Valley (n = 268 boxes), we developed a nest box selection model and evaluated the model’s predictive performance in other regions and years. We hypothesized the model would perform better in regions with similar crops and landcover than in areas more ecologically dissimilar. We found that the top model performed well for predicting future nest box occupancy in Napa, but it performed poorly in ecologically similar Sonoma, and in dissimilar Fresno, Madera, and Merced Counties. These results suggest models predicting barn owl nest box occupancy from a single region may not be generalizable to others. Instead, ecologists should consider the need to develop region-specific models to better understand barn owl habitat ecology and the ecosystem services they provide to farmers.